1) Genomic instability. Genomic instability is undoubtedly a hallmark of many malignancies, and one might guess that malignant cell lines would display increased genomic instability. We have recently begun to study mutation frequency in leukemic cell lines, using an assay for mutation of the X-linked hprt locus. We find that although some cell lines have dramatically increased mutation frequencies (100x that of normal peripheral blood T-lymphocytes), others have decreased mutation frequency with respect to T-lymphocytes. Through molecular analysis of the hprt mutants, we were able to determine that individual cell lines showed either gross chromosomal rearrangements (GCR) or point mutations, but not both, with rare exceptions. Of interest, two of three cell lines which display point mutations but not GCR are know to be mismatch repair deficient. We have begun to extend these findings by assaying the types (GCR or point mutations) of hprt mutations present in lymphoblastoid cells of patients with DNA repair/checkpoint lesions, such as Bloom's syndrome, ataxia-telangiectasia, and Fanconi's anemia. 2) Recombinogenic genomic regions. Despite thousands of studies based on recurrent, non-random chromosomal translocations, a fundamental question concerning these translocations remains unresolved. This question can be phrased as follows: "Do recurrent, non-random translocations occur between "recombinogenic" regions of the genome that are extraordinarily susceptible to breakage/religation events, or are the regions involved not particularly recombinogenic, but simply regions that lead to the production of oncogenic fusion proteins which confer a growth advantage to the cell." To help address this question, we are developing an in vitro system in which we can produce translocations, through the use of the Isce-I restriction endonuclease, that do not necessarily provide a growth advantage to the cell.